Prostate cancer cells that survive androgen deprivation are a potential source for androgen-independent CaP. Recurrent CaP shares many phenotypic characteristics with the purported prostate epithelial stem cells suggesting a possible linear evolution from undifferentiated prostate stem-like cell, though a tumor stem cell with arrested differentiation and resistance to androgen deprivation-induced apoptosis, to recurrent androgen-independent CaP. Therefore, the hypothesis to be tested in this proposal is that a stem-like cell in the prostate undergoes genetic changes and gives rise to a malignant tumor stem cell that proliferates in response to androgen, resists apoptosis upon androgen deprivation, and is the source of androgen-dependent CaP. The experimental plan to test this hypothesis is divided into two strategies: Specific Aims 1 and 2 will focus on the prostate stem cell as a target for neoplastic transformation and Specific Aims 3 and 4 will evaluate the role of the prostate tumor stem cell in the transition from androgen-dependent to androgen-independent growth. Specific Aim 1 will isolate and characterize stem-like cell lines from the regenerating rat prostate. The goal of these studies is to isolate stem-like ells at different stages of differentiation, contrast the expression of basal cell and secretory cell specific genes in these cell lines, end evaluate the capacity of these stem-like cells to differentiated into mature prostatic epithelial when co-transplanted with normal prostatic stromal cells or when transplanted into the urogenital sinus. Specific Aim 2 will derived and characterize a cohort of androgen-dependent and androgen- independent prostatic epithelial tumor cell lines derived from a subset of the stem-like cell lines established in aim 1. The goal of these studies is to generate spontaneous neoplastic transformants; characterize their expression of the basal cell and secretory cell specific markers; isolate androgen-independent variants of the tumor cell lines; and determine if expression of the androgen receptor is required to maintain tumorigenic potential. Specific Aim 3 will compared the tumor stem cell components of the androgen-dependent Dunning H Tumor and CWR-22 tumors. The goals of these studies are to establish the methodology to identify and characterize the tumor stem cells in xenografts, establish tumor stem-like cells in culture, and compare their phenotype to recurrent versions of the xenografts. Based upon the methodologies and markers established in aims 1-3, Specific Aim 4 will characterize the stem cell and tumor stem cell components in histologic sections of androgen-dependent and androgen-independent human prostate cancers and in transient xenografts to be established from normal human prostate and CaP. The goals of these studies are to determine if a tumor stem cell compartment exists in human CaP, if it could contribute to the evolution of recurrent CaP and if there are racial differences in the stem cell and tumor cell compartments between African Americans and Caucasians.